Error when using delayed writing using one area but several (writers) formats #3285
Description
Describe the bug
I get an error when trying to make AVHRR images using the delayed writing functionality (as is used in trollflow2). See the example code below. The original error was discovered when running trollflow2 with a the minimal productlist, see extract outlined below:
reader: avhrr_l1b_aapp
mask_area: true
delay_composites: true
use_tmp_file: true
radius_of_influence: 10000
metadata_aliases:
variant:
GDS: global
EARS: regional
DR: direct_readout
min_coverage: 25
areas:
euron1:
areaname: euron1
products:
overview:
min_sunlight_coverage: 10
productname: overview
formats:
- format: tif
writer: geotiff
blockxsize: 512
blockysize: 512
tiled: true
- format: png
writer: simple_image
- format: jpg
writer: simple_image
fill_value: 0
...
If comment out either the png and jpg and leave only the tiff, it works fine. If I comment out the tiff, and leave the png/jpeg it also works fine!
To Reproduce
I managed to reproduce it with this minimal code below:
# Your code here
scene = Scene(filenames=filenames, reader='avhrr_l1b_aapp')
parameters = ['overview']
scene.load(parameters)
AREAID = 'euron1'
local_scn = scene.resample(AREAID)
start_time_txt = local_scn.start_time.strftime('%Y-%m-%d %H:%M')
timestamp = local_scn.start_time.strftime('%Y%m%d_%H%M')
obj1 = local_scn.save_dataset('overview', writer='geotiff',
blockxsize=512,
blockysize=512,
tiled=True,
compute=False,
filename='avhrr_ovw_%s_%s.tif' % (timestamp,
AREAID))
obj2 = local_scn.save_dataset('overview', writer='simple_image', fill_value=0,
compute=False,
filename='avhrr_ovw_%s_%s.jpg' % (timestamp,
AREAID))
obj3 = local_scn.save_dataset('overview', writer='simple_image',
compute=False,
filename='avhrr_ovw_%s_%s.png' % (timestamp,
AREAID))
objs = [obj1, obj2, obj3]
compute_writer_results(objs)
Expected behavior
I would expect three images being written, now I get only two, the png and the tiff.
Actual results
Text output of actual results or error messages including full tracebacks if applicable.
Traceback (most recent call last):
File "/home/myuser/test_satpy_save_images.py", line 73, in <module>
compute_writer_results(objs)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^
File "/path/to/conda/env/lib/python3.13/site-packages/satpy/writers/core/compute.py", line 139, in compute_writer_results
da.compute(delayeds)
~~~~~~~~~~^^^^^^^^^^
File "/path/to/conda/env/lib/python3.13/site-packages/dask/base.py", line 681, in compute
results = schedule(expr, keys, **kwargs)
File "/path/to/conda/env/lib/python3.13/site-packages/pyresample/future/resamplers/nearest.py", line 62, in query_no_distance
distance_array, index_array = kdtree.query(
~~~~~~~~~~~~^
coords,
^^^^^^^
...<2 lines>...
distance_upper_bound=radius,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
mask=mask)
^^^^^^^^^^
File "pykdtree/kdtree.pyx", line 274, in pykdtree.kdtree.KDTree.query
ValueError: Mask must have the same size as data points
Screenshots
If applicable, add screenshots to help explain your problem.
Environment Info:
- OS: Linux (RHEL9)
- Satpy Version: 0.57.0
- Pyresample Version:
- bokeh: 3.8.0
- trollimage=1.26.0
- Readers and writers dependencies (when relevant): [run
from satpy.utils import check_satpy; check_satpy()]
See requirements file attached!
Additional context
Add any other context about the problem here.